Summer running - 2000 Ducati ST2

[OddDuck]

Brass sprocket shim - gone! Completely destroyed in around 1,000 km of riding. The fragments photographed next to the allen key set are the biggest bits left. I got some flakes and fragments off the inside of the sprocket cover but really that's it. I'd expected it to get chewed up sooner or later but seeing it completely absent was a surprise. OK, so not everything works out - the brass really is too soft for this. It might be worth trying again with slightly thicker stainless, but I'd want a more assertive method of folding it into spline grooves than pushing with a ruler, and that means machining up some kind of custom rolling tool. At this point it's getting into more serious territory than this particular output shaft deserves.

Earlier I'd machined up an external cover, similar to the collar but fitting over the lock washer and its bolts. It axially restrains the front sprocket and helps stop tilt but nothing more, it won't stop it rattling back and forth. Torsional lock onto the output shaft is provided by machining it slightly undersize, so that the splines force grooves into it on fitting (these can be seen in the photo if you look closely). For now I'll run with this and the collar and see how it works out. There's some chain lube in the spline teeth, too. The moly grease from earlier is almost completely gone as well, it seems to flow and fling, and without O-ring seals or similar I think more will just disappear in short order.

[OddDuck]

I've replaced the ignition relay. Just a very short note about this...

The photo shows the new relay (the cover is sealed), the relay I took off the ST2 in the center, and a slightly older relay off the 900SS on the right. The ST2's former relay is clearly on the way out. The solenoid coil is discoloured and it also had that smell of cooked electrical insulation. The contactor pads looked alright though.

Two of the relay's underside terminals are discoloured, apparently by sustained heating. The other terminal blades appear alright. This is concerning - local heating like this would imply that there's a resistance in the crimp connectors which push on here, or heat coming down out of the relay itself. I haven't checked the crimp terminal connectors in the loom itself yet.

Relay types: it turns out that 87 and 87a (marked on the underside) do mean different things, replacing with the correct type is the difference between the bike firing up good as gold or absolutely nothing happening when the key is turned. It's the only difference between the two relay types. I've marked my spares as Injector or Ignition so I don't get them confused later.

I took the bike for a ride with the new relay and there's a slight further improvement. Still not perfect but liveable for now.

[neels]

I've replaced the ignition relay. Just a very short note about this...

The photo shows the new relay (the cover is sealed), the relay I took off the ST2 in the center, and a slightly older relay off the 900SS on the right. The ST2's former relay is clearly on the way out. The solenoid coil is discoloured and it also had that smell of cooked electrical insulation. The contactor pads looked alright though.

Two of the relay's underside terminals are discoloured, apparently by sustained heating. The other terminal blades appear alright. This is concerning - local heating like this would imply that there's a resistance in the crimp connectors which push on here, or heat coming down out of the relay itself. I haven't checked the crimp terminal connectors in the loom itself yet.

Relay types: it turns out that 87 and 87a (marked on the underside) do mean different things, replacing with the correct type is the difference between the bike firing up good as gold or absolutely nothing happening when the key is turned. It's the only difference between the two relay types. I've marked my spares as Injector or Ignition so I don't get them confused later.

I took the bike for a ride with the new relay and there's a slight further improvement. Still not perfect but liveable for now.

Might have to replace the relays on my ST2 on suspicion......

Relay terminal 87 is the NO contact, 87a is the NC contact, so if they are reversed the relay is working the wrong way around.

Had a similar problem with the fuel pump relay in my son's mercedes, relay contacts looked fine but after a clean it went back to starting every time, not just when it felt like it.

[OddDuck]

Tried a quick experiment on the weekend - a back country loop that incorporated 30 km of gravel roads. Hilly, hard base, not really a heavy street bike's natural environment. I'd bought a small bicycle pump with a pressure gauge and tried deliberately deflating the tyres from the normal 38 psi to 25 (ish) to improve grip. We re-inflated at the other end to highway pressure.

The reasoning for this was that transitions on short stretches of gravel are a reality if you're back country touring in NZ. It's nowhere near as frequent as it used to be: the network of fully sealed roads expands every time the local councils have spare budget to get rid of. The sealed portions lengthen and finally link up, closing loops. However any really interesting touring will involve gravel at some point. It might just be a couple of k's, it might be more. So it pays to be prepared to ride it.

Anyway, the stretch that I tried was about the best test possible for the idea: hilly, narrow, up and down, pitched cambers and washboard corrugations, and loose stones on a hard base.

The tyre deflation didn't really work as well as I'd hoped. I just ended up with more tyre sitting on top of gravel, which then rolled between rubber and the road base. Expanding the contact patch via deflation helped slightly but it was no substitute for knobblies. Sorting inflation at each end was a hassle; highway pressures on a hot tyre were also inaccurate and the tyres needed topping up again later on. The other issue with this technique (I found this out talking with people after trying it) was the very real risk of tyre delamination due to flexing and overheating. This could lead to catastrophic failure at speed on the sealed highway later on.

In short, dual purpose tyres exist for a reason.

The rear brake proved twitchy, too. The ST series bikes are known for a lack of feel on the rear - it's either on or off, without much in the way of progression. On a loose surface like gravel, running slow, the rear brake is generally best to use - if there's a lockup then that's a lot easier to recover from than the front going sideways abruptly. It took a couple of goes before I learnt to go super light on the rear. Going downhill though, even a moderate slope effectively meant near to zero ability to stop the bike.

The other thing that happened was overheating. I'd had a few ideas about what it'd be like taking the ST2 onto gravel but hadn't expected this: the bike really didn't cope with extended running at under 30 km/h. There simply wasn't enough airflow through the radiator, even with the engine only running at light throttle. Halfway through, the 120+ flashing warning came up and I had to close down for twenty minutes. If I'd been able to post a 45-ish average (on that surface) I'd have been alright, but I'm simply not that good a rider. That was reinforced quite nicely by a relatively mild camber putting me into the ditch... it only took about 15 degrees sideways camber to do it too. Once in, it's hard to get the bike back out again, with the rear wheel skidding like hell on the edge of the ditch and the whole bike going sideways. It was bad but better than what could have happened if I'd met an oncoming vehicle and been unable to pull over to the left properly. There were way too many corners where I was in the middle or even right hand side of the road, simply because of the cambers.

After the ditch I realised why the handful of vehicles I'd seen had been rolling so slowly (well under 50 k's): even the farmer's utes and SUVs were having significant tyre drift. Apparently there's also a legal issue with these types of roads, the law here is that you need to be able to stop in half the distance that you can see ahead of you. That could mean being found at fault should there be some sort of collision and liable for the other guy's repairs. So ability to go slow counts.

It was a very useful learning experience but I'll try to keep the ST2 on seal only from now on.

[neels]

I've ridden my ST2 on gravel, it's a pig.

[noelh]

I have actually ridden and do ride my ST2 on gravel LOTS, yes it is a pig and i have been in the ditch and done tank slappers a few times but in reality it is no different to a mulitstrada or other big dual purpose bike, One thing i have noticed (compared to the Pirelli angels I had prior) is the Michelin road 5s i have, as well as being much better in the wet work really well on gravel because of the big grooves in them. (and they work really well on the road too, they are all i buy now).

I have never had any overheating problem either, and we do 30-40 km metal roads sometimes so not sure why you had a heating issue?

I can generally pass most dual purpose bikes because the ST2 being so low actually sits better than a taller bike, reminds me of my XR500 30 years ago.

Theres my 10c worth...

As far as the main relay goes, my one has 2 x 87 terminals and a 30 terminal, no 87a so it is not a changeover type, I probably should have mentioned that. The non change over ones often have 2 x 87 terminals and if going into a 4 pin socket you break off the not required pin with long nose pliers.

I use silicon grease on my terminals (except the Main EFI computer connector, dont EVER grease that multi pin connector)

Keep riding and keep the info coming.

[OddDuck]

The overheating's probably from riding like a great big scaredy-cat, I hate that feeling of the bike being loose so I slow right down. More gravel experience on something lighter and much cheaper to drop would probably sort that out.

Thanks for the tip about the Michelin Road 5's. I'm running Bridgestone T31 Evos and so far have been very happy with them on tarmac both wet and dry, however they're almost slicks in the middle and terrible on gravel.

Overheating - it still isn't quite where it should be - I can see three further possibilities:

1) I've got another high resistance spot somewhere in the wiring loom between coils and battery, believable on a 20 y.o. bike

2) The exhaust gaskets are starting to let go and the engine is sucking outside air into the cylinder on the transition between exhaust pulse and induction cycle, running lean as a result (it has been sounding different lately)

3) A sensor has gone bad and is telling the CPU wrong things about (say) air temperature

Checking 3) is going to be tricky but 1) can be done with a thermal imaging camera once the fairing's off and 2) is a teardown and visual check. I've already got the spares on order (the header nuts and studs are a rusted solid mess) so hopefully I can get this done in the next few days.

[neels]

3) A sensor has gone bad and is telling the CPU wrong things about (say) air temperature

Have had this one with the water temp sensors, when the fairings were off I noticed that the wiring looked a bit odd, the cables had been swapped between the front and rear cylinder sensors. Turns out one connects to the ECU, the other to the temp gauge. Put them back where they belonged and temp gauge was intermittent due to crook solder joints on the sensor connector pins, hacked it open and added a short cable between the sensor and connector and all good again.

Radiator fan on mine comes on at about 100° on the temp gauge, this is controlled by the ECU via the above sensor.

[OddDuck]

Thanks Neels - I've replaced both water temp sensors around a year ago so I'm sure these are OK.

Having had spare parts arrive, I took the mufflers and the headers off today and had a nosey. The bottom two fairing panels had to come off - so far so good - then I had to unscrew the corroded on M8 steel nuts on the exhaust headers. Standard procedure is to soak them in PB Blaster and leave to wick in for around two hours, then use a six-point socket. The two on the horizontal cylinder were easy enough, as was the left hand side of the vertical cylinder, but I had a hell of a time with the vertical exhaust's inboard nut. It's tucked right into all the clutter behind the engine and also into the curve of the header pipe. I broke it loose with a ring spanner, then had to undo the thing about a twelfth of a turn at a go with the open end. That was all the swing I had to work with due to the clutter. A stubby would have been very helpful and a set of these are now on the shopping list. I just couldn't get anything in my socket systems to fit.

It's possible that a 1/4" format socket, universal joint and extension bar might work, but the real problem here was the use of standard M8 nuts. They're 13mm across the flats, the sockets of course are bigger, and even thin-walled 3/8th format just doesn't fit into that header pipe's shoulder. Ducati use a special copper alloy nut with a 10mm hex and this would have been one hell of a lot easier to work with.

It kept on like this... there's gasket goo on the clamp joint between headers, stainless fasteners screwed into plain steel on the mufflers, and I'm pretty well certain that the old exhaust gaskets were re-used. They've been crushed right down.

A look inside the exhaust ports, and at the gasket faces of the header pipes, confirmed a minor exhaust leak on the vertical cylinder. So it was worth getting in to sort this out.

Stainless fasteners on the mufflers may be a problem due to galvanic corrosion. They're contacting the aluminium outer shell of the mufflers, and are threaded into something on the inside made of what looks like a grade of mild steel. The steel inner component has rusted, badly, and one of the stainless bolts galled its threads on being unscrewed, locked, and broke off. It didn't take much force to break it, either. I suspect that the stainless hasn't been handling repeated heat-cool cycles very well, even though it hasn't visibly degraded. I'll be attempting to drill and extract this later. OEM Ducati fastening here is a passivated steel M8 hex head bolt and what looks like a special large diameter (possibly aluminium) washer, I'll probably return to this or similar. BZP is possible but I'm not sure how well the zinc will behave once threaded into plain steel and everything heated to whatever temperature the mufflers get to normally.

This is a problem because (as I relearned again) reassembly of the Ducati muffler system depends on hanging everything back onto the bike - headers, mufflers, all brackets etc - with all fasteners loose. It can't be done in stages because it's too easy to set the headers at an angle and then not feed into the mufflers properly. So everything's got to be in place first before tightening of any fasteners.

As to why I was so keen to get this bolt undone... whoever worked on the mufflers previously has used badly undersized washers, which have cut their way into the aluminium hangar brackets. The edges of the washers are engaged and are cutting into the bracket. They're slowly but surely forcing the mufflers to move, which will be putting a force onto the header cross joint. This may be a factor in why the exhaust system started leaking, too.

The photos show the headers and ports as they were when everything came off the bike. I've cleaned everything up and reinstalled headers with fresh gaskets and proper OEM Ducati header nuts, I'll have to attempt to sort the muffler out later.

[OddDuck]

I got the broken off and jammed in bolt out yesterday. This stuff is the mechanic's nightmare... the head's gone, the bastard thing is jammed tight, there's not enough stub left to get vice grips onto it, and there's no way it can be ignored. The usual story is drill and apply an EZ-Out; I don't think this would have worked here. Once threads gall together, there's a ball of metal-to-metal weld and any attempt to force it just snowballs that weld. You can force the thing out (if the EZ-Out doesn't either slip or snap) but you'll rip the thread out while doing it. I didn't have an EZ-Out small enough anyway.

I accidentally found a technique which might be worth sharing. Most of this isn't in the photos unfortunately.

1) drill right through the bolt, use a small diameter pilot drill at first

2) progressively expand the drill diameter until the hole wall is nearly at the threads, it will almost certainly be off-center but that's alright as long as there's a paper-thin wall on the old bolt

3) use a pin punch around the broken stub to smash the fastener inwards, make it collapse into itself via the hole through the center

4) it should now be loose, if pliers or EZ-Out can unscrew the thing then job done

5) if it jams, progressively smash the old bolt inwards by pin punch on the fastener walls, centerpunch flat areas and use a small drill to gradually remove metal and work down

6) once it's down a few threads in the hole, the original female thread can be used as a drill guide. Caution with this though - if it's a steel fastener into aluminium there's a very good chance that the drill will walk sideways into the threads instead of staying on-axis and on-center on the old bolt.

7) run a tap through the threads and clean out all fragments prior to fitting a new fastener. A single shaving left in a thread can cause the new bolt to lock up like the old one. A compressed air gun with a nozzle fine enough to go to the bottom of the hole is the ultimate for cleaning out a tapped thread in a blind hole, but cotton buds, white spirits and patience works too.

I had a couple of advantages with this particular job: there was enough stub left to fit a standard M8 nut and use that as an initial drill guide, and it was threaded into thick-walled steel, not soft aluminium. The pin punch smash-up technique might be too brutal for alloy, particularly engine casings. A lot of care is needed with the drill while doing this, it'd be very easy to break small drill bits with sudden break-throughs and drills binding up, or just pushing too hard. You really need to be in a non-angry, non hurry sort of a mood when doing this kind of work despite how easy it'd be to get frustrated.

One thing that I should mention: stainless steel work hardens very quickly during drilling if it's allowed to get hot. Initially the shavings come out and things look good, then the drilling slows right down, then drills go blunt very quickly. If you've got smoke coming off things then it's already too hot. Speed and force don't help here, keeping the material cool does. The ideal is to have a machinist's water-based cutting fluid and keep it wet at all times. I didn't have this so had to resort to a jam jar filled with water and frequent dips of the drill into this, plus lots of aerosol cutting oil. It's messy but it works. Re-sharpening drills helps greatly as well, this can be done fairly quickly with a standard bench grinder and a bit of practice.

In the end the old fastener came out as a pile of drill shavings. There wasn't a recognisable fragment of the old bolt left. I was very pleased to find that despite the abuse, the old thread was in nearly perfect condition. There'll be no issue with using the muffler again.

The tap was very tight when cleaning out the threads, which is why I think the jam happened in the first place. Time and oxidation have taken their toll here: steel expands when it rusts so tight clearances close up. One of my little rules is that a fastener should twirl into place between thumb and forefinger. If it won't then something is wrong.

[Laava]

Jeez bro, you sure are patient!
Something you need to be with Ducati ownership as I am regularly reminded...

[OddDuck]

Haha thanks Laava!

Yeah there were a couple of moments where I could have gone full Hulk and smash smash smash... but no. It was a challenge.

Just a really quick note about further work today... the ST series use a muffler with a swivel joint, this means that the muffler is pinned in position to the headers instead of being able to move along a slide joint, and that means that the hangar bracket needs to move along the axis of the muffler when refitting. This has to be done each time when changing exhaust gaskets etc since very small movements or angles at the cylinder head means large movements at the swivel. So the two mounting bolts actually need to loosen and retighten occasionally, you can't set them once and that's it.

After a think about the issues, I've settled on this solution for now:

- BZP grade 8.8 fasteners and washers, not stainless

- use of copper coat anti-seize on threads

- positively retaining the head of the bolt, not the thread, against vibration and loosening.

I've done the last by a spring washer and an M8 mudguard washer which has been curved to match the curve of the hangar. That means there's a positive shape match between the two, it isn't just a flat washer sitting and being held by friction. The spring washer should ideally dig into both the bolt and the washer, just a bit, so there's the positive retention required. The mudguard washer was bent by holding it in a vise (soft jaws used) and hitting it with a hammer. The three lines show the positions in the vise jaws. It's a bit rough but so far seems alright, although I'll be checking fasteners very closely after each ride for a while.

[OddDuck]

The bike's developed a really nasty tingle through the bars and the pegs lately - feels like detonation or mis-timed ignition. It seems to be a part of the overheating issues as well. It's a funny effect though, the bike's smooth like it should be first thing, good for about twenty-ish minutes, and then the effect sets in and just gets worse if I keep riding. I've been having a look and reckon there are three areas of concern:

1) old rubber in the cooling system, especially the O-rings connecting to the cylinder heads
2) old rubber on the inlet manifolds after the throttle plates, again O-rings to the cylinder heads
3) old crimp connectors in the wiring loom, especially around the battery.

The ignition depends on being able to use the battery as a snubber; it's where pulses from ignition coils get smoothed out. Good clean connections are essential. I can (and have) cleaned up battery terminal posts, spade lug faces, etc, but of course it's not possible to get inside a crimp terminal. I've spent the last couple of nights having a good look at the originals, both main loom and starter motor cables, and it looks like they're getting a bit tired... hopefully the attached photos show a few things. I took the time to check the 40 amp rec-reg fuse as well since the connectors were showing their age, but thermally this came in OK.

The thermal images were taken with a FLIR C2 mixed image thermal / optical camera of the positive and negative battery terminals and cables, immediately after an engine start. It's a cheaper unit with an offset between the two imaging systems, hence the apparent mismatch during close-up work. The bit of brass shim wrapped over the steel battery terminal adaptors was an earlier attempt to improve things, which actually did help (especially with starting) but didn't stop the problem occurring once the bike was running and warmed up.

Just a really quick note about the thermal camera - for diagnostic work it is a very helpful tool. I could clearly see high resistance spots in the loom, or confirm that apparently old connectors were still OK. It wasn't easy to get good photos but a lot could be seen just walking around the bike with it. The other thing was that I'd been concerned about the cooling system, both thermostat and impellor. The thermal camera clearly showed the entire radiator quickly warming up once the bike warmed up to nominal thermostat opening temperature.

[OddDuck]

A further look at the crimps... I've had a go at making my own starter cables, engine ground, and replacing the main loom's battery crimps. As part of that I did some destructive testing, namely cutting test crimps in half through the crimp area and polishing the ends up.

Ideally the crimp should have a closed perimeter and no gaps anywhere inside. It should be 100% clean metal throughout, any oxides or oils will get trapped in the joint. The hexagonal shape is the new work I've done. The closed letter-C shapes, in brass on copper wire, are the OEM Ducati crimps, in this case the first sectioned item with black insulation is one of the starter cables, the second (with red and purple in the background) is the loom connection to the positive terminal on the battery.

There are vacancies clearly visible on the Ducati crimps, gaps between the wire bundle and the crimp terminal, and also a hexagonal pattern of oxides between wire strands. I suspect that the brass has been relaxing with time and opening up, allowing oxides to work their way up into the crimp zone and start to break it up. Maybe the oxides forced their way in, in a similar effect to steel bodywork going rusty under paint a long way from the open damage, not that it matters now. It looks a great deal like it wasn't crimped particularly hard at the factory. The connector isn't in good shape, in other words.

My new work isn't perfect (there are still vacancies in the wire strand bundle) but they do look like an improvement. That's the starter cables and engine ground. Unfortunately I've had to put a new terminal over old wiring for the loom connectors. The ground side of the loom was still good shiny copper but the positive side had gone brown... I've taken the risk and crimped over dull copper. In hindsight I should have pickled the wire beforehand with some jeweller's Pre-Pro, it's good stuff available online through Regal Castings. I've used this before on annealed copper washers with success, it dissolves all oxides right down to bare copper. We'll see if this works, if it doesn't I guess I'm into replacing wiring or trying to re-crimp again.

For those interested in the super close up photos and how they were done... 35mm true macro lens capable of 1:1 magnification, plus 25mm extension tube. The lens had a ring light which was very helpful.

[OddDuck]

Some progress today, replaced old starter cables with new, applied heatshrink over the loom connectors, and made a new pair of brass battery terminal adaptors to replace the steel Ducati originals.

The new battery terminal adaptors are made from cut-down premium marine battery terminals. I got these off Burnsco for $12 each. It took about half an hour per each with a hacksaw, file, drill and M6 tap to get them to fit to the bike.

First start has been tried and the bike's starting, and cold idling, has immediately improved. While the engine was running I took the chance to shoot a couple of thermal images, front and rear of the battery, showing the positive and negative terminals. Cable heating appears greatly reduced but it's not quite an apples to apples test; ambient on the previous test was around 18 C and today it's in the mid-20's. The cables look like they're running cooler relative to background but that's all I can be sure of.